{"id":"W4224247167","doi":"10.1101/2022.04.19.488797","title":"High resolution shotgun metagenomics: the more data, the better?","year":2022,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"Graduate and Professional Student Council, University of Arizona; Compute Canada","keywords":"Metagenomics; Shotgun sequencing; Workflow; Deep sequencing; DNA sequencing; Sample (material); Computer science; Shotgun; Data mining; Computational biology; Biology; Genome; Database; Genetics; Gene","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008494563,0.0008851624,0.00123272,0.001040753,0.0007894209,0.003842098,0.001464158,0.001721797,0.001377898],"category_scores_gemma":[0.008556563,0.0003278239,0.001123957,0.00288359,0.001245447,0.003900102,0.001518256,0.00316218,0.0005990173],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001175592,"about_ca_system_score_gemma":0.001153851,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003144974,"about_ca_topic_score_gemma":0.006030923,"domain_scores_codex":[0.9958548,0.002018711,0.0002440283,0.0007815475,0.000868483,0.000232323],"domain_scores_gemma":[0.9921867,0.003335488,0.0005758,0.001265889,0.002209876,0.0004262214],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003404079,0.0008841823,0.1565785,0.006344872,0.002236788,0.001882459,0.001946099,0.06950264,0.3996909,0.02514828,0.04560822,0.2867731],"study_design_scores_gemma":[0.0002484366,0.001114371,0.1027138,0.00211568,0.001410032,0.001164036,0.008160448,0.2278731,0.3025443,0.1411139,0.2109374,0.00060437],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.7423696,0.02128814,0.1879546,0.02616007,0.001458914,0.0002147497,0.01200294,0.001725383,0.006825625],"genre_scores_gemma":[0.7904429,0.00652301,0.1814789,0.005420336,0.0005356083,0.0001840481,0.01371377,0.0004099838,0.001291362],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.008494563,"threshold_uncertainty_score":0.04492414,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02069220652782459,"score_gpt":0.2308375629512859,"score_spread":0.2101453564234613,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}